Three major catalysts are expected to boost the relative performance of Chinese assets

Core Viewpoints

Although the nation celebrated the Spring Festival last week and the domestic market began its holiday on Tuesday, the global market experienced a tumultuous week. The domestically developed "extreme cost-performance" large model DeepSeek R1 made a stunning debut in the global market; within ten days of taking office, Trump issued 67 presidential orders and announced a 25% tariff on Mexico and Canada, causing a resonance in the stock, bond, and foreign exchange markets, but the extent of tariffs on China and the overall policy stance were milder than expected; additionally, the risk management of Chinese real estate companies took a significant step forward. We believe that these three major changes could serve as important catalysts for the relative performance of asset prices in China and even globally. Huatai Macro has also provided commentary on each (see "The Macro Impact of U.S. Tariffs on Mexico and Canada"; "Will DeepSeek Shake the Investment Logic of the Stargate" (2025/2/3); "Some Positive Changes in China's Real Estate Industry Recently" (2025/2/2)). This short article summarizes the uplifting effect of these three events on the relative valuation of Chinese assets. Indeed, as the global market still needs to digest the impact of the U.S. officially announcing tariffs over the weekend (February 1), short-term volatility may rise, but the relative performance of Chinese assets may still have support.

1. The U.S. Announces Tariffs on Initial Countries, with Tariff Levels on China Lower than Expected

Within ten days of taking office on January 20, President Trump issued 67 presidential orders and memoranda, and on January 31 (Friday), he announced that tariffs on Mexico and Canada would be implemented earlier on February 1, at a rate of 25%. At the same time, on February 1, Trump signed a presidential order announcing a 10% tariff on Chinese export goods. This series of changes undoubtedly impacted the dollar, U.S. Treasury bonds, and the U.S. stock market, and since the announcement coincided with the weekend, the effects of these policies are still unfolding (see "The Macro Impact of U.S. Tariffs on Mexico and Canada," 2025/2/3).

However, it is noteworthy that from the signals in Trump's inauguration speech on January 20 (see "Changes and Constants After Trump's Inauguration," 2025/1/21) to yesterday (February 1, U.S. time) regarding the extent and wording of tariffs on the first batch of the top three countries with trade surpluses with the U.S., the hawkishness of the Trump 2.0 administration's policies related to China, including tariff policies, has been milder than market expectations. Additionally, a series of signals from interactions between the two heads of state after January 20 indicate that the Trump administration is still seeking cooperation with China in various fields, including geopolitical issues, investments in the U.S., and trade rebalancing.

If the short-term trajectory of U.S.-China relations is more positive than the previously pessimistic market expectations, the corresponding geopolitical "discount" in the valuation of Chinese assets, especially those listed overseas, will narrow. It is not ruled out that before the formal meeting between the two heads of state (after April), U.S.-China relations may enter a window period. It is well known that the strategic competition pattern between the U.S. and China and the U.S. tariff policy towards China are significant driving factors for the valuation of Chinese assets, especially those listed overseas. Between 2018 and 2019, the U.S. imposed multiple rounds of tariffs on China, and each round of tariff increases was accompanied by an increase in the risk premium of Chinese assets, including those listed in Hong Kong and the U.S Offshore RMB exchange rate, etc. (Charts 1 and 2). However, if the possibility of a phased easing in Sino-U.S. relations increases, not only will the impact of tariffs on Chinese corporate profits be less than expected, but the corresponding valuation discount of Chinese assets may also narrow.

2. Real Estate Market and Risky Property Companies Release Positive Signals

After the Politburo meeting in September last year proposed "promoting the stabilization of the real estate market," multiple ministries introduced a "combination punch" of policies to boost real estate transactions.

Looking ahead to 2025, even under pessimistic assumptions, the direct drag of the real estate industry on the economy this year is expected to converge—this is more due to changes in the supply-demand structure and improved affordability following market adjustments. However, some positive factors, including the stabilization of real estate transactions, are accumulating recently. The year-on-year growth rates of national new and second-hand housing transactions recorded 10%/27% from October to December last year. Although there was a pullback in transaction growth after the pulse-like rebound brought by the real estate tax incentives in December, filtering out the disturbances from the Spring Festival shift, the real estate transaction activity in January remained high, especially with second-hand housing transactions continuing to record an 18% year-on-year growth (Charts 3 and 4). The speed of tightening supply and demand varies across regions—cities with strong fundamental support, such as population inflow and rapid economic development, show significantly stronger momentum and sustainability in the rebound of real estate transactions compared to other cities. Admittedly, further macro policy efforts may still be needed to drive the real estate market towards further recovery. However, the early signs of stabilization and rebound in real estate transactions in some cities are expected to marginally alleviate some investors' concerns about the rapid slowdown in domestic demand growth. In addition, it is worth noting that the recent management changes at Vanke Group, which serves as a "barometer" for risk disposal in the real estate sector, have released a somewhat positive signal. On one hand, the entry of Shenzhen state-owned enterprise ShenTie into Vanke has alleviated buyers' concerns about Vanke's ability to deliver homes. At the same time, this move has also significantly mitigated the risks of uncertainty in real estate companies' debt disposal being transmitted to the financial system. As a result, on the day the management change announcement was made, the interpretation of related assets was relatively positive (Charts 5 and 6). Certainly, considering that the cash flow of many real estate companies still needs to be repaired, the solvency of real estate companies, including Vanke, still needs further improvement, and the sustainability of current policy effects also requires further verification.

3. DeepSeek Attracts Global Market Attention

In our view, the hottest investment theme during the Spring Festival remains a series of discussions surrounding the successful "comeback" of the Chinese local enterprise DeepSeek. From January 25-26, as DeepSeek R1 topped the U.S. app download charts, this Chinese startup instantly attracted the attention of major media and websites, becoming a hot topic in the financial market and among the general public (Chart 7).

Why has DeepSeek attracted global attention?

• The success of DS has shaken the absolute monopoly expectations of the U.S. in the forefront of AI large model research and development, and has also, to some extent, broken the myth that large model development can only achieve "miracles" with ultra-luxury hardware/chips/resources support.

• The relative valuation gap in the China-U.S. AI industry chain is expected to marginally converge. On one hand, the popularity of DS has ignited market enthusiasm for investing in Chinese software and computing infrastructure companies; on the other hand, the "alternative possibility" provided by DS has also shaken the current investment logic of the U.S. AI industry chain to some extent.

• On a macro level, DS has, to some extent, impacted the monopoly excess profits (rent) represented by Nvidia and the "Tech Seven" of the U.S. AI industry chain, and the transfer of these excess profits to numerous software development companies and users will help improve overall social productivity (see "Will DeepSeek Shake the Investment Logic of Stargate?") Compared to major American companies, DeepSeek's hardware conditions and R&D costs are significantly constrained. However, the "extreme cost-performance ratio" large model launched by DeepSeek, such as the inference model R1, can perform comparably to the most advanced similar models in the U.S. in mainstream evaluations, but at only about one-twentieth of the cost. At the same time, DeepSeek adopts an open-source model, making it possible for small teams to develop at low costs and high levels, raising doubts about the long-term competitiveness of the currently high-cost, high-consumption development model in the U.S. On January 27, the "DeepSeek" effect triggered a deep adjustment in the U.S. technology sector, with the Nasdaq index falling by 3.1% that evening, NVIDIA's stock price dropping by 17%, and a market value evaporation of about $500 billion—equivalent to the total market value of the Mexican stock market, just from one stock in one day.

Therefore, DeepSeek's success 1) narrows the perceived gap in the development prospects of the AI field between China and the U.S. to some extent; 2) is expected to ignite the investment enthusiasm of Chinese internet companies in large model development, applications, cloud computing, and other AI infrastructure areas; 3) disperses the monopolistic profits currently concentrated in the U.S. AI hardware and large model development fields; 4) the emergence of high cost-performance large models is conducive to promoting the popularization of AI applications domestically and even globally (i.e., the process of transforming from "rent" to "utility").

Recently, the release of the DS large model has accelerated, and it is expected that other Chinese internet companies will also quickly launch their independently developed AI models. On January 29, Alibaba Cloud's Tongyi Qianwen flagship model Qwen2.5-Max was officially released. According to Alibaba Cloud's official website, the Qwen2.5-Max model is the latest achievement of the Tongyi team on the MoE model, with pre-training data exceeding 20 trillion tokens. The new model demonstrates strong overall performance, achieving high scores on multiple public mainstream model evaluation benchmarks.

Thus, DS not only prompts a rethinking of the development stages of the China-U.S. AI industry chain but also will lead to a repricing of the investment logic and profit distribution patterns in the AI industry chain. Currently, there is still a significant valuation gap between the Chinese AI industry chain and U.S. companies across various segments, including development, hardware, software, and applications, with the proportion of the total market value still significantly low (Charts 8-11).

![](https://mmbiz-qpic.wscn.net/mmbiz_png/bLkoW2zgquyNEIE7N2wQic0MkicNicVcib5dQzic8TCR2tmKvdmLgL66yzzdJyIOAiayxV5uj7mtgoMicEULOHaUyqakw/640? Overall, the recent increase in the probability of a "window period" in Sino-U.S. relations and substantial progress in the risk management of the real estate industry are expected to significantly reduce the "tail risk" that has suppressed the valuation of Chinese assets over the past three years; meanwhile, the success of DeepSeek indicates that Chinese private enterprises still possess strong vitality and creativity—during the era of Industry 4.0, China has broad prospects for expanding at the technological frontier, applying new technologies, and developing new markets.

In addition, although the foundation for stabilizing and recovering domestic demand still needs further consolidation, there have been some positive changes in China's real estate market since the fourth quarter of last year (see "Recent Positive Changes in China's Real Estate Industry," 2025/2/2). Among them, recent substantial progress in blocking the transmission of risks from the real estate deleveraging cycle to the financial system helps to reduce the tail risk of the Chinese cycle and the risk premium of assets. This article summarizes a series of unexpected changes in industry, cyclical trends, and foreign relations, as well as their implications for the relative performance of Chinese assets.

Risk Warning: Further escalation of global trade frictions; unexpected downturn in the real estate cycle dragging down domestic demand.

Author of this article: Yi Han, Source: Huatai Securities Research Institute, Original Title: "Three Catalysts Expected to Boost the Relative Performance of Chinese Assets."

Yi Han Researcher SAC No.S0570520100005 | SFC AMH263

Risk Warning and Disclaimer

The market has risks, and investment requires caution. This article does not constitute personal investment advice and does not take into account the specific investment objectives, financial conditions, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their specific circumstances. Investment based on this is at one's own risk